DE69709169T2 - PRODUCTION OF VINYL ACETATE - Google Patents

Description

The Government of the United States of America owns rights in this invention under Cooperation Agreement No. DE-FC22-95PC93052 awarded by the US Department of Energy.

The invention relates to a process for producing vinyl acetate by contacting a mixture of hydrogen and ketene with a transition metal-containing heterogeneous catalyst to produce acetaldehyde, which is then reacted with ketene in the presence of an acid catalyst to produce vinyl acetate. The acetaldehyde used in the present invention is produced by the process set forth and claimed in U.S. Application Serial No. 08/619,385, filed March 21, 1996, the contents of which are incorporated herein by reference.

Background of the invention

Vinyl acetate is an important industrial chemical. Most of the vinyl acetate used in industry is polymerized to poly(vinyl acetate). This important polymer and its derivatives are widely used as adhesives, paints and other coatings, films and laminating materials.

Vinyl acetate has been produced commercially by reacting acetylene with acetic acid at 180-210ºC in a vapor phase process. Although yields based on acetylene and acetic acid typically exceed 90% for both reactants, the high cost of acetylene, as well as the safety and handling problems associated with its use, make this process disadvantageous when compared to ethylene-based processes.

An example of an ethylene-based vinyl acetate process uses acetaldehyde and acetic anhydride as starting materials. Ethylene is used to produce acetaldehyde by the Wacker oxidation process, and the resulting acetaldehyde is with the acetic anhydride in the presence of an acid catalyst to produce ethylidene diacetate (EDA). The EDA is then reacted in the presence of an acid catalyst to produce vinyl acetate, acetic acid, acetic anhydride, and acetaldehyde. This EDA cracking reaction is undesirable for at least two reasons: 1) the byproduct acetic acid is formed, which must then be converted back to acetic anhydride or otherwise used or disposed of, and 2) unfavorable equilibria exist among EDA, the desired products (vinyl acetate and acetic acid), and the undesirable original starting materials (acetaldehyde and acetic anhydride). Thus, the EDA cracking reaction must be carried out in the presence of a large excess of acetic anhydride to increase the amount of vinyl acetate and acetic acid produced relative to acetaldehyde. However, even if excess acetic anhydride is added to the EDA cracking reactor, the product distilled from the reactor still contains high levels of acetaldehyde, acetic anhydride and acetic acid in addition to the desired vinyl acetate. Therefore, this process requires multiple distillations and extensive recycling. Because of these problems, coupled with the corrosion and safety and handling problems associated with the Wacker process, it is more economical to produce vinyl acetate by the direct reaction of ethylene, acetic acid and oxygen.

Currently, the preferred route to vinyl acetate is the direct reaction of ethylene, acetic acid and oxygen to form vinyl acetate, water and by-products. The preferred version of this process uses a heterogeneous catalyst and is carried out in a vapor phase at 5-10 bar (500-1000 kPa) at 150-200ºC. Because of the explosion risks associated with this reaction, the reaction must be carried out with less than a stoichiometric amount of oxygen; thus, the conversions of ethylene, acetic acid and oxygen are typically 10-15%, 15-30% and 60-90% respectively. Approximately 5-10% of the ethylene is converted to carbon dioxide and approximately 1% is converted to acetaldehyde. The low ethylene and acetic acid conversions per pass require extensive recycling along with a carbon dioxide removal system. Although the capital cost of an ethylene-acetic acid-oxygen based vinyl acetate plant is high, this capital cost is offset by the generally low cost of ethylene and acetic acid. Therefore, there is a need for a process for producing vinyl acetate with higher conversions per pass and a lower yield loss to carbon dioxide than the ethylene-acetic acid-oxygen route. Oxygen-based. The process of the present invention, in contrast to the ethylene-acetic acid-oxygen-based route, actually produces vinyl acetate at high conversions per pass and does not produce significant amounts of carbon dioxide.

Other attempts to produce vinyl acetate have also been made. For example, a number of these processes attempt to produce vinyl acetate from mixtures of carbon monoxide and hydrogen (syngas) due to the very low cost of the raw materials. As a first step, these systems convert syngas to methanol or dimethyl ether. In addition, numerous combinations have been tried in which methyl acetate (produced from methanol and recycled acetic acid) or dimethyl ether is carbonylated to form acetic anhydride. In some schemes, acetic anhydride is partially hydrogenated to form EDA and acetic acid. In still other systems, the methyl acetate or dimethyl ether is carbonylated in the presence of hydrogen to form EDA and acetic acid in one step. Variations on this approach include reacting methanol or methyl acetate with hydrogen and carbon monoxide to form acetaldehyde and water or acetaldehyde and acetic acid; however, the selectivity to acetaldehyde in these reactions is poor. The resulting acetaldehyde is then reacted with acetic anhydride to form EDA.

None of these all-syngas routes are economically competitive with the current commercial ethylene, acetic acid and oxygen-based process. Since the production of vinyl acetate by a syngas-based process ultimately involves the cracking reaction of EDA, the process is hampered by many of the same operational difficulties of the above-mentioned EDA process based on ethylene-derived acetaldehyde and acetic anhydride. Furthermore, most all-syngas routes require that two moles of acetic acid be recycled or used in another way. The methanol hydrocarbonylation route is theoretically somewhat advantageous in that the only byproduct, acetic acid, is formed in the EDA cracking step; however, this advantage is offset by the poor selectivity to acetaldehyde that characterizes these reactions.

As mentioned above, fully syngas-based processes were based on acetic acid formed as a by-product by EDA cracking, which must be recycled or used in another way or disposed of. Therefore, there is a need for a process for producing vinyl acetate which does not have the feature of recycling large amounts of acetic acid which is characteristic of EDA cracking based processes. Since the process of the present invention produces vinyl acetate directly from acetaldehyde and ketene without a concurrent significant formation of acetic acid, the process of the invention does not have a significant acetic acid recycling problem and does not require a large scale acetic anhydride production step.

The process of the invention has not been described previously. Unlike the present invention, most of the previous processes using acetaldehyde as a feedstock also use acetic anhydride as a feedstock to produce EDA, which is subsequently cracked to acetic acid and vinyl acetate. U.S. Patent No. 2,425,389 teaches that the preferred catalysts for cracking EDA to acetic acid and vinyl acetate are aromatic hydrocarbon sulfonic acids and that the cracking reaction should be carried out in the presence of at least a three-fold molar excess of acetic anhydride over EDA to control the balance to acetic acid and vinyl acetate. The process of U.S. Patent No. 2,425,389 also teaches that acetic anhydride must be continuously added to the mixture to replace that which is depleted.

In Hydrocarbon Processing 44 (11), 278 (1965), the process described for the continuous production of vinyl acetate from acetaldehyde and acetic anhydride teaches that acetaldehyde and acetic anhydride are reacted together in a reactor to form EDA; the EDA is then cracked in a separate reactor tower to form vinyl acetate and acetic acid, which are removed as overhead product together with acetaldehyde. A series of distillations is carried out to separate and recover vinyl acetate, acetic acid and acetaldehyde. The acetaldehyde is then recycled to the EDA reactor. The process of the present invention, in contrast, does not require the continuous introduction of acetic anhydride and produces at most only trace amounts of acetic acid when carried out under the preferred conditions.

Brady teaches in The Chemistry of Ketenes, Allenes and Related Compounds, Part 1, S. Patai (editor), John Wiley and Sons, New York, 292 (1980), that ketenes and aldehydes in presence of Lewis acids to form beta-lactones. Thus, Japanese Patent Application No. 47-25065 teaches that ketene and acetaldehyde react at 5-15°C in the presence of the Lewis acid boron trifluoride to form beta-butyrolactone, and compares the activity of this catalyst with those of other Lewis acids such as zinc chloride and iron tetrafluoroborate. Japanese Patent Application Nos. 49-131718 and 49-134954 teach that silica-alumina catalysts (which possess both Lewis and Bronsted acidity) are also active catalysts for the conversion of ketene and acetaldehyde to beta-butyrolactone at 10-15°C. Thus, the prior art deviates from the teachings of the invention because when ketene and acetaldehyde are reacted in the presence of Lewis acids or the solid acid silica-alumina, which has both Lewis and Bronsted acid character, the result is betabutyrolactone.

It is well known that ketenes can react with enolizable ketones in the presence of Bronsted acids to form enolic esters at 50-100°C as described, for example, in U.S. Patent No. 2,487,849. Furthermore, Japanese Patent Application No. 48-75510 states that the efficiency of the reaction of ketene with enolizable carbonyl compounds using the catalyst system described therein is related to the ease of enolization of the carbonyl compound. In particular, in the presence of the Bronsted acid-based catalyst system of Japanese Patent Application No. 48-75510, ethyl acetoacetate reacts with ketene more readily than acetone to form the corresponding enol esters. Furthermore, March teaches in Advanced Organic Chemistry, 4th ed., John Wiley and Sons, New York, 585 (1992) that the rate of acid-catalyzed enolization is proportional to the concentration of protonated carbonyl present. In the same reference, March demonstrates on page 250 that it is approximately 1000 times more difficult to protonate an aliphatic aldehyde than to protonate an aliphatic ketone with an arene sulfonic acid. Therefore, the prior art teaches that it should be very difficult to prepare an enol ester from an aliphatic aldehyde.

U.S. Patent No. 2,422,679 describes the reaction of ketenes with aldehydes to form unsaturated carboxylic acid esters in the presence of a strong Bronsted acid catalyst in a temperature range of 0 to 80°C. U.S. Patent No. 2,422,679 strongly suggests that strong acid catalysts such as sulfuric acid are preferred over weaker acids such as p-toluenesulfonic acid, which is consistent with the teachings of U.S. Patent No. 2,487,849, Japanese Patent Application No. 48-75510 and those of March above. Even when sulfuric acid is used as a catalyst, the efficiency of the process of U.S. Patent No. 2,422,679 is low; and when about 5.4 moles of acetaldehyde are reacted with 2 moles of ketene (bubbled through the acetaldehyde-sulfuric acid mixture at 15°C), only about 21% of the ketene is converted to vinyl acetate after the mixture is distilled. (Note that the figures given in Example 1 of this patent are more consistent with ketene conversion to vinyl acetate of about 21% than with acetaldehyde conversion of 21%; the conversion of acetaldehyde to vinyl acetate is calculated to be about 8% based on the figures of Example 1.) The process of the present invention is a clear improvement over the process of U.S. Patent No. 2,422,679, since higher yields of both ketene and acetaldehyde are obtained in the present invention by using higher temperatures and weaker catalysts under continuous conditions rather than the lower temperatures and stronger catalysts under batch or semi-batch conditions in U.S. Patent No. 2,422,679.

European Patent Application 0 348 309 A1 describes a process for converting EDA to vinyl acetate in the presence of a ketene stream and an acid catalyst (preferably sulfuric acid). The ketene in the process reacts with acetic acid formed from EDA cracking to produce acetic anhydride; as the acetic anhydride level builds up, there is a modest increase in the amount of vinyl acetate produced compared to that produced in the absence of ketene. However, in the process of European Patent Application 0 348 309 A1, much of the ketene is wasted and the formation of vinyl acetate decreases as EDA is depleted. Furthermore, the EDA used in the process of European Patent Application 0 348 309 A1 must be prepared in a separate step before the reaction can proceed. The process of the present invention is much more efficient than that of European Patent Application 0 348 309 A1, since a separate EDA preparation step is not required; in the present invention, ketene and acetaldehyde are fed into a reactor together and the ketene input is much higher.

Summary of the invention

The present invention provides a more efficient and economical route for producing vinyl acetate than previous processes. As mentioned above, the present invention obviates many of the disadvantages of the prior art. For example, the The process of the present invention does not use significant amounts of carbon dioxide and gives higher conversions per pass than the ethylene-based processes. In addition, the present invention avoids the high byproduct formation of acetic acid which complicates the synthesis gas-based processes. In addition, the process does not require the continuous introduction of acetic anhydride. Further advantages of the present invention are both set out above and apparent from the examples set out below.

The present invention relates to a process for producing vinyl acetate comprising the steps of contacting a mixture of hydrogen and ketene with a transition metal hydrogenation catalyst to form acetaldehyde and then contacting a mixture of the resulting acetaldehyde and ketene with an acid catalyst to form vinyl acetate. More particularly, the present invention relates to a continuous process for producing vinyl acetate comprising the steps of continuously feeding a gas comprising ketene and hydrogen and optionally a non-reactive diluent gas into a first contact zone containing a transition metal hydrogenation catalyst; continuously removing acetaldehyde from the first contact zone; continuously feeding a gas comprising ketene and the acetaldehyde from the first contact zone and optionally a non-reactive diluent gas into a second contact zone containing an acid catalyst and optionally a solvent; and continuously recovering a product comprising vinyl acetate from the second contact zone. Our new process for producing vinyl acetate has numerous advantages over the prior art processes as set forth above, including more efficient production of vinyl acetate and improved yields.

Detailed description of the invention

The aim of the present invention is to provide an efficient process for forming vinyl acetate. In the present invention, hydrogen and ketene are reacted in the presence of a hydrogenation catalyst to form acetaldehyde. The resulting acetaldehyde is further reacted with ketene in the presence of an acid catalyst to form vinyl acetate.

The process of the invention can be carried out in two separate contact zones or it can be carried out in a single contact zone. When two separate contact zones are used, a first contact zone is used to hydrogenate ketene to acetaldehyde in the presence of a transition metal catalyst; a second contact zone is used to react the acetaldehyde formed in the first contact zone with ketene in the presence of an acid catalyst to produce vinyl acetate. When a single contact zone is used, ketene and hydrogen are reacted in the presence of a transition metal catalyst and an acid catalyst to produce vinyl acetate. It is generally preferred to use two separate contact zones because, in the case of a single contact zone being used, the reduction in the acid catalyst will reduce its activity and inhibit the activity of the transition metal catalyst. If the reaction continues to occur in a single contact zone, the acid catalyst can be present as either a soluble catalyst or an insoluble solid acid catalyst; if the acid catalyst is an insoluble solid acid catalyst, the degradation of the acid catalyst and the inhibition of the transition metal catalyst can be slowed. However, if two contact zones are operated, a mixture of hydrogen, ketene and acetaldehyde exiting the first contact zone can be fed into a second contact zone to produce vinyl acetate without significantly affecting the activity of the acid catalyst.

The ketene used can be prepared by any of the usual ketene-forming reactions such as acetic acid dehydration, acetic anhydride pyrolysis, diketene pyrolysis or acetone pyrolysis. On an industrial scale, it is preferred to prepare the ketene by acetic acid pyrolysis for economic reasons.

The acetaldehyde used in the present invention is prepared by hydrogenating ketene according to the process described in U.S. Application Serial No. 08/619,385, filed March 21, 1996. Thus, the resulting acetaldehyde should be free of large amounts of components that normally react with ketene. That is, it is preferable to have acetaldehyde that is substantially free of nucleophilic impurity components such as water or alcohols, since these can react with ketene, resulting in a loss of yield. It is often advantageous to use trace amounts (greater than 5 ppm) of a polymerization inhibitor such as copper or a hydroquinone derivative in ranges of the process where vinyl acetate is present, but this is not a requirement of the invention.

The acetaldehyde used in the present invention is prepared by hydrogenating ketene under very mild conditions. The process can be used in combination with known ketene preparation processes to convert a variety of acetyl and related compounds, such as acetic acid, acetic anhydride, diketene and acetone, to acetaldehyde. The preparation of acetaldehyde in the present invention is carried out by the steps of (1) contacting hydrogen and ketene gases with a catalyst comprising a metal selected from the elements of Groups g and 10 (IUPAC classification; Group 9 = Co, Rh and Ir; Group 10 = Ni, Pd and Pt) of the periodic table in a hydrogenation zone and (2) recovering acetaldehyde from the hydrogenation zone. As mentioned above, if two separate contact zones are used to produce vinyl acetate, the hydrogenation of ketene to acetaldehyde would take place in a first contact zone, with the product from this being able to be fed directly into a second contact zone. Our new process for producing acetaldehyde does not involve the formation of ketene-metal complexes of the type described in the literature.

The process for producing acetaldehyde can be operated at temperatures ranging from 0 to 250ºC, although low temperatures give low reaction rates and excessively high temperatures cause accelerated degradation resulting in loss of yield. Therefore, a more preferred range of temperatures is between 50 and 200ºC. The most preferred temperature range is between 70 and 150ºC.

The catalytic hydrogenation process for producing acetaldehyde can be carried out at pressures in the range of 0.05 to 100 bar (5 to 10,000 kPa) absolute (the pressures referred to herein are bar absolute and kPa). However, excessively high pressures increase the possibility of forming ketene polymerization products, whereas excessively low pressures result in lower reaction rates and it is difficult to remove the heat from the reaction. The process is preferably carried out at a pressure of 0.1 to 20 bar (10 to 2000 kPa), with the most preferred range being 0.25 to 10 bar (25 to 1000 kPa). Since ketene is normally produced at a pressure of 1 bar (100 kPa) pressure or lower, the hydrogenation is suitably carried out at a pressure of 1 bar (100 kPa) or lower.

The reactant mixture in the hydrogenation zone (i.e., the first contact zone) may consist essentially of 100% ketene and hydrogen, or a non-reactive (inert) diluent gas such as nitrogen, argon, helium and light hydrocarbon may be added. For example, the presence of non-reactive diluent gas in the reactant mixture may assist in the removal of heat from the reaction zone. When used, inert diluents may comprise 1 to 95 volume percent of the reactant feed. The use of excess amounts of diluent gas reduces the reaction rate and makes isolation of the product acetaldehyde more difficult. The presence of significant amounts of carbon monoxide may adversely affect the hydrogenation catalysts, particularly the preferred palladium catalysts. Therefore, the reactant mixture should normally contain less than 1% by volume carbon monoxide, preferably less than 1000 ppm carbon monoxide.

The hydrogen to ketene molar ratio can vary considerably and can range from 0.25:1 to 10:1. The hydrogen/ketene molar ratio is preferably in the range from 1:1 to 8:1, most preferably 2:1 to 4:1. Hydrogen/ketene molar ratios below 1:1 limit the conversion of ketene and reduce the reaction rate. Although the reaction rate increases with increasing hydrogen/ketene ratios, excessive amounts of hydrogen increase the difficulty encountered in isolating the product. Also, the use of excessive amounts of hydrogen combined with a low space velocity can result in the formation of some ethanol or ethyl acetate after most of the ketene has been consumed. Ethanol is not normally formed in the process of the invention, but ethyl acetate is detectable at higher conversions.

The metals which catalyze the hydrogenation of ketene to acetaldehyde according to the present invention are found in a formally designated Group VIII or Group VIIIA of the Periodic Table and in particular in the presently designated Groups 9 and 10 of the Periodic Table. The catalyst is preferably selected from rhodium, platinum and in particular palladium. The catalytic metals may be used in the form of unsupported metals or they may be used in the form of a supported catalyst comprising the catalytic metal deposited on a catalyst support material. Alumina, carbon, titanium dioxide, silica, Barium sulfate, barium carbonate, and calcium carbonate are examples of suitable support materials. The Lindlar catalyst (palladium on calcium carbonate modified with lead) is also effective in the reaction, but is not as selective as palladium on the other supports mentioned above. When a support is used, the metal loading can range from 0.1 to 10 wt.%. Metal loadings outside these ranges will also carry out the reaction, but generally do not optimize the use of the metal and support. It is also preferred to use an unsupported palladium catalyst, such as sponge palladium, because hydrogen treatment often restores activity more effectively than is the case with a supported catalyst. It may also be possible to use the catalytically active metals in the form of salts or complexes that are soluble in a liquid reaction medium in which the process can be carried out.

Various modes of operation can be used in the production of acetaldehyde according to the invention. For example, the process can be used as a heterogeneous vapor phase process in which a vapor (gas) comprising ketene, hydrogen and optionally a non-reactive diluent gas is fed to a reaction (hydrogenation) zone containing one or more beds of the catalysts described above. An alternative heterogeneous mode of operation consists of a vapor/liquid/solid phase process in which a feed gas comprising ketene, hydrogen and optionally a non-reactive diluent gas is fed to a reaction zone containing the catalyst as a finely divided suspension in a non-reactive liquid reaction medium such as mineral oil. The product acetaldehyde can be removed from the reaction zone by gas stripping. In another embodiment of the vapor/liquid/solid phase mode of operation, a mixture of the gas feed and a non-reactive liquid can be fed into the hydrogenation zone where it is passed over the solid catalyst in a trickle bed mode of operation. Finally, the process can be practiced using a homogeneous catalyst solution consisting of a salt or complex of the catalytically active metal dissolved in a non-reactive liquid reaction medium (solvent) to which the gas mixture comprising ketene, hydrogen and optionally a non-reactive diluent gas is fed. Such homogeneous operation is not preferred.

The hydrogenation process can be operated as a batch, semi-continuous or continuous process. The most efficient operation of the hydrogenation process is achieved by continuously operating the process in a heterogeneous gas phase mode of operation. In this preferred mode of operation, the process of the invention provides for a continuous production of acetaldehyde by the steps of:

(1) Continuously feeding a vapor (gas) comprising ketene, hydrogen and optionally a non-reactive diluent gas into a reaction (hydrogenation) zone containing one or more beds of the catalytically active hydrogenation catalyst; and

(2) continuously removing a product gas comprising acetaldehyde from the reaction zone.

The catalysts used in the preferred heterogeneous continuous process include supported and unsupported palladium catalysts. As mentioned above, in the present invention, the product acetaldehyde can then react with ketene in the presence of an acid catalyst to form vinyl acetate.

With respect to the hydrogenation reaction, the gas hourly space velocity (GHSV - volumes of reactant per volume of catalyst per hour) of the ketene-containing reactant and the feed of diluent gases can range from 10 to 100,000 using the preferred heterogeneous mode of operation. The GHSV is preferably in the range of 100 to 50,000, and most preferably in the range of 1,000 to 20,000. In general, an increase in GHSV increases the reaction rate but decreases the conversion. The selection of the optimal GHSV depends on the physical form of the catalyst and the rate and conversion desired.

As mentioned above, when the process of the invention is carried out within two contact zones, ketene and the acetaldehyde from the first contact zone and an acid catalyst are contacted in a second contact zone to form vinyl acetate. The second contact zone may contain acid catalyst in a liquid solution as a solid or as a mixture of solid and liquid. The preferred acid should be a Bronsted acid. More preferred acids are those containing phosphorus or sulfur in a positive oxidation state and these include the liquid phosphorus, Sulfuric and methanesulfonic acids, the soluble solid benzenesulfonic, p-toluenesulfonic, naphthalenesulfonic and naphthalenedisulfonic acids and an insoluble acidic ion exchange resin such as Amberlyst® 15 (a partially cross-linked sulfonic acid form of polystyrene) and polymeric Nafion® 117 (a perfluorinated sulfonic acid resin) sulfonic acids. Strong acids such as sulfuric acid cause excessive charring under the reaction conditions and can form the byproduct acetic acid. Weak acids such as methanesulfonic acid give lower reaction rates. Polystyrene-based catalysts such as Amberlyst® 15 tend to decompose slowly under the reaction conditions. Volatile acids such as methanesulfonic acid tend to leave the product in the reaction zone, which is undesirable. The even more preferred acids are arene sulfonic acids: benzenesulfonic acid, p-toluenesulfonic acid, naphthalenesulfonic acids, and naphthalenedisulfonic acids. In general, p-toluenesulfonic acid is the most preferred acid because no significant amounts of acetic acid are produced after steady state operation is achieved.

When the second contact zone contains acid catalyst in liquid solution, the acid concentration should be in the range of 0.005 to 2.0 acid equivalents per liter of solution. The process of the invention is possible with acid concentrations outside this range, but excessively thin acid concentrations tend to provide lower rates and excessively high acid concentrations tend to cause excessive charring. The preferred acid concentrations depend on the acid used, the identity of the solvent, the contact time and the temperature. Thus, for example, if the acid is p-toluenesulfonic acid, the solvent is acetic anhydride, the space velocity is about 65 hr-1 (units are liters of ketene + acetaldehyde at reaction temperature and pressure per liter of solution contacted per hour, excluding the volume of any diluent), and the temperature is 150°C, a more preferred range of acid concentration is 0.01 to 1.2 acid equivalents per liter of solution, and the most preferred range for acid concentration is 0.03 to 0.8 acid equivalents per liter of solution.

If the second contact zone contains an insoluble solid acid, it should contain 0.01 to 5 acid equivalents per liter of catalyst. Preferably, the solid insoluble acid should contain 0.1 to 3 acid equivalents per liter of catalyst, and most preferably between 0.5 and 2 acid equivalents per liter of catalyst. After the solid insoluble acid is contacted with After the ketene and acetaldehyde reactants in the second contact zone have been contacted under the reaction conditions, a liquid phase is formed and the operating mode then changes to a gas-liquid-solid mode.

When the second contact zone contains acid catalyst in liquid solution, it is preferred that the acid, ketene and acetaldehyde have good solubility in the solvent. The solvent may or may not react with ketene or acetaldehyde to form reactive intermediates. Interactions of the solvent with ketene or acetaldehyde should be such that any intermediate species formed are ultimately converted to vinyl acetate. Ideally, the solvent should boil at a temperature equal to or higher than the reaction temperature. Examples of classes of solvents suitable for the reaction include, but are not limited to, polycyclic aromatic hydrocarbons such as naphthalene and their halogenated derivatives such as 1-chloronaphthalene, polar aprotic solvents such as N-methyl-2-pyrrolidinone, acetic anhydride, ethylidene diacetate and mixtures of these. Byproduct liquids from the reaction can be used as solvents or solvent components. The solvents can contain small amounts of water or acetic acid. The water and acetic acid react with the ketene early in the reaction and are largely converted to acetic anhydride. The preferred solvents for the reaction are acetic anhydride or acetic anhydride containing small amounts (10 mole % or less) of acetic acid. When these solvents are used with the preferred catalysts, EDA is formed early in the reaction and builds up to a constant level (approximately 20 to 40 weight % depending on temperature), and the reaction rate increases as the level of EDA builds up and remains relatively constant thereafter. However, when N-methyl-2-pyrrolidinone is used as the solvent, the level of EDA present in the solution remains at about 1 weight %. In general, when water is present in the solvent or associated with the acid catalyst, the efficiency of the reaction increases as the water is consumed.

After the reaction rate in the second contact zone has reached its maximum, the mass of liquid solution contained in the contact zone remains essentially constant, except for the slow formation of non-volatile by-products and entrainment or vaporization of the solvent. The entrainment or vaporization of the solvent is minimized by the use of a distillation column to these components back to the contact zone solution while allowing vinyl acetate and any unreacted ketene or acetaldehyde to leave the contact zone. Slowly draining a portion of the liquid from the lower portion of the second contact zone and simultaneously replenishing it with fresh liquid by methods known to those skilled in the art can prevent accumulation of excess non-volatile by-product.

The process of the invention can be carried out at pressures in the range of 0.05 to 20 bar (5 to 2000 kPa) absolute in the second contact zone. However, excessively high pressures increase the possibility of ketene polymerization reactions and, by reducing product volatility, can make it more difficult to continuously recover the vinyl acetate product in those cases where the product must evaporate from a liquid medium. In contrast, excessively low pressures reduce reaction rates, can complicate heat control and make condensing and recovering the vinyl acetate product more difficult. The process is preferably carried out at a pressure of 0.1 to 5 bar (10 to 500 kPa) in the second contact zone, with the most preferred range being 0.25 to 2 bar (25 to 200 kPa). In addition, since ketene is normally produced and used at a pressure of 1 bar (100 kPa) or less, the process of the invention is most conveniently carried out at a pressure of 1 bar (100 kPa) or less.

Diluent gas may be fed with the ketene and acetaldehyde into the second contact zone, and this may assist in removing the desired vinyl acetate product from the contact zone. However, excess amounts of diluent gas may increase the difficulty of condensing and recovering the vinyl acetate from the vapor phase. Diluent gases may comprise 0 to 95 volume percent of the ketene-acetaldehyde feed mixture. More preferred proportions of diluent gases are in the range of 0 to 90 volume percent, and most preferred proportions of diluent gases are in the range of 0 to 85 volume percent. The diluent gases used should be those that do not react with the reactants or products under the reaction conditions. Suitable diluent gases include hydrocarbons such as methane, nitrogen, helium or other inert gas, hydrogen and gases commonly present in ketene streams including, in addition to methane, carbon monoxide, propyne, allene, ethylene, acetylene and carbon dioxide. Air or other molecular oxygen-containing gases are not preferred diluent gases. due to safety considerations, although the vinyl acetate producing reaction occurs when these gases are used as diluents.

The ratio of ketene to acetaldehyde in the second contact zone is not critical and can range from 0.1 to 10. If a large excess of any of the reactants is used, the unreacted portion must either be recycled to the reaction contact zone or used in another way. If ketene is the excess reactant, it can be recycled or converted to other useful products, such as acetic anhydride or diketene. If acetaldehyde is the excess reactant, it can be recycled, recovered unchanged, or converted to other useful products, such as pentaerythritol, synthetic pyridine derivatives, or peracetic acid. A more preferred ratio of ketene to acetaldehyde is between 0.2 and 5 and the most preferred ratio of ketene to acetaldehyde is between 0.4 and 2.5.

The temperature of the second contact zone can range from 85ºC to 200ºC. Excessively low temperatures will result in low rates and conversions, whereas excessively high temperatures will accelerate the rate of char formation, increase catalyst losses and result in little rate increase. More preferred temperatures are in the range of 100 to 180ºC, and the most preferred temperature range is 120 to 160ºC when the contact zone contains acid catalyst in liquid solution. Insoluble, solid Amberlyst® acid catalysts decompose at elevated temperatures and should not be used at temperatures much above 120ºC, but the Nation® catalysts can be used at temperatures as high as 150ºC.

The second contact zone should be designed to allow efficient contact with the acid catalyst, allow sufficient time for the reactants to react, be capable of accepting non-volatile components recycled to the reactor, and operate without significant heat or mass transfer limitations. The actual configuration of the second contact zone may depend on the scale of operation, flow rates, temperature, pressures, proportion of diluent, nature and amount of solvents present, and catalyst type and amount. In most cases, the acetaldehyde and Ketene is fed near the bottom of the second contact zone along with any diluent gases. The simplest example of a second contact zone for an acid catalyst in a liquid solution is a cylindrical tube with a provision to spray gases into the lower part of the tube.

In some cases, it may be preferable to circulate the liquid medium countercurrent to the reactant gas streams to improve heat and mass transfer in the second contact zone. The catalyst-containing liquid may be dispersed over a high surface area solid if desired to improve gas/liquid contact, but some degree of liquid circulation is required to replenish the solid surface with fresh catalyst solution. The use of baffles, high speed stirrers or other gas dispersing devices to improve mass transfer is within the spirit of the invention.

When an insoluble solid acid catalyst is used in the second contact zone, the reaction can be carried out without the addition of liquid. However, generally in this case a liquid will appear at the end and wet the catalyst. The insoluble solid acid can be supported in a tube and the reactants can be passed through the lower part of the tube or through the upper part of the tube, and generally the reactor operates in a trickle bed mode. Alternatively, the insoluble solid acid catalyst can be dispersed in a liquid and the reaction can be carried out in a slurry mode, with the reactants and suspensions being handled in the same manner as for soluble catalysts and using similar liquid components.

It is best to choose the rate of reactants through the second contact zone after other operating parameters have been decided. This rate can vary considerably depending on the other operating parameters and must be adjusted experimentally to provide optimum conversion of the limiting reactant and the highest selectivity and rate. Under operating conditions of 0.03 equivalents of arsenic sulfonic acid per mole of acetic anhydride solvent at 130-160ºC and with approximately 85% diluent gas present, between 5 and 160 liters of acetaldehyde plus ketene vapor are passed per liter of catalyst solution per hour. neglecting the volume of diluent gas. More preferred feed rates under these conditions are in the range of 30 to 120 liters of acetaldehyde plus ketene vapor per liter of catalyst solution per hour neglecting the volume of diluent gas. Most preferred feed rates under these conditions are in the range of 45 to 80 liters of acetaldehyde plus ketene vapor per liter of solution per hour neglecting the volume of diluent gas.

The process of reacting ketene and acetaldehyde can be operated as a continuous process as described above or, less preferably, as a batch process. For example, a batch process may require that the desired amounts of reactants, solvents and catalysts be charged to a contact zone, heated under pressure until the reaction is complete, and then the vinyl acetate is recovered by distillation. Batch operation is not preferred because prolonged heating of the reactants and products in the presence of the acid catalyst results in the formation of large amounts of byproducts. Thus, the process is preferably operated as a continuous process as described above. Due to the short time for which the reactants, products and acid catalysts are contacted at an elevated temperature, the formation of undesirable byproducts is minimized when the process is operated in the continuous manner described above. After the continuous process reaches a steady state power condition, the rate of mass discharge into the contact zone approximately equals the rate of mass discharge from the contact zone. Under the preferred operating conditions, acetic acid is not present in the product stream in significant amounts after a steady state power condition is reached.

Examples General experimental methods

For a review of the apparatus used in the hydrogenation of ketene to acetaldehyde, see U.S. Application Serial No. 08/619,385, filed March 31, 1996, the contents of which are incorporated herein by reference. In addition, the application contains examples to further illustrate the preparation of acetaldehyde used in the process of the invention.

An overview of the apparatus used in Examples 1-11 of this application is as follows. The apparatus consisted of 7 main sections:

1. a gas supply system;

2. a ketene production section,

3. a ketene purification section,

4. a ketene capture/evaporator section,

5. an acetaldehyde purification and feeding section

6. a reactor cooler series section,

7. a scrubber system.

These 7 main sections were connected by a series of connecting pipes and stopcocks. The gas supply system delivered four separate metered nitrogen streams to different points in the apparatus. The ketene generation section consisted of an arrangement of an acetic anhydride vaporizer/pyrolysis tube, a cooled condenser and a cyclone.

produced crude ketene by acetic anhydride pyrolysis and provided rapid separation of the ketene from the bulk of the byproduct acetic acid and unreacted acetic anhydride. The ketene purification section consisted of a defogging trap (trap A) maintained at 0ºC and a trap (trap B) initially maintained at -78ºC for the initial condensation and subsequent distillation of purified ketene to the ketene trap/evaporator section. The ketene trap/evaporator section served as a reservoir for purified ketene and also provided for the metered feed of ketene to the reactor. The ketene trap/evaporator section was maintained at -78ºC for the collection of distilled ketene from trap B and its subsequent feed to the reactor by transpiration in metered nitrogen. The acetaldehyde purification and feed section consisted of a graduated cylinder from which acetaldehyde could be distilled and a trap/evaporator section maintained at -20ºC for collecting the distilled acetaldehyde and then feeding it to the reactor by transpiration in metered nitrogen. The reactor cooler series section consisted of a reactor where the ketene and acetaldehyde reacted, a water cooled cooler to return condensable liquid to the reactor and an acetone/solid carbon dioxide cooler and trap (trap C) to collect the bulk of the vinyl acetate, vaporized acetic acid and to collect unreacted acetaldehyde. The scrubber system included a water scrubber for the degradation of ketene and an analytical scrubber which used acetic acid for the quantification of ketene and for the capture and quantification of any vinyl acetate or acetaldehyde not captured by Trap C of the reactor-cooler bank section.

The apparatus was designed so that the reactants could bypass the reactor as an option. The apparatus also had a provision for a continuous flow of nitrogen to prevent back diffusion of scrubber fluids or contamination by air. Details of the construction of each section of the apparatus, how they are connected, and the operation of the apparatus are described below.

In the gas supply section, metered nitrogen gas flows were provided by four Tylan® Model FC-260 mass flow regulators. Metered nitrogen gases were supplied by 4 gas lines L1 through L4, and each gas line was connected to a water-containing pressure equalization column via a tee to prevent accidental overpressurization. Two of these gas lines, L1 and L3, provided nitrogen purge to the device. L1 provided nitrogen purge to the acetic anhydride pyrolysis unit, and L3 provided nitrogen purge to the rest of the device at a point connected via a tee to the ketene trap/evaporator vent line described below. Gas lines L2 and L4 were used to dose ketene and acetaldehyde from the ketene trap/evaporator section and the acetaldehyde trap/evaporator section, respectively, by transpiration.

In the ketene generation section, ketene was produced by the pyrolysis of acetic anhydride by the process described by Fisher et al. in J. Org. Chem., 18, 1055-1057 (1953) with minor modifications (although acetic pyrolysis is the preferred industrial route for ketene, it is generally not practical on a laboratory scale). Acetic anhydride feed was provided by a Harvard apparatus, Model 22 syringe infusion pump. The acetic anhydride was fed into the top of a vertical quartz evaporator/pyrolysis tube, 107 cm long and 25 mm OD, along with nitrogen metered in through line L1. Electrical temperature control for the evaporator/pyrolysis tube and monitoring were provided by a Dow Camile® control system interfaced with a Gateway Model 2000 486DX/33- Computer. The evaporator/pyrolysis tube was notched 27 cm from the top and contained a 9 cm OD quartz thermocouple extending well over two-thirds of the length of the reactor from top to bottom. The portion of the evaporator/pyrolysis tube extending 22 cm up from the notches also contained quartz chips and was heated with a heater band set at 200ºC. The lower section of the evaporator/pyrolysis tube was heated by two heater bands set at 520ºC. The quench cooler beneath the evaporator/pyrolysis tube was maintained at about -55ºC by circulating methanol cooled in an acetone/solid carbon dioxide bath. The mixture from the quench cooler was passed sequentially through two identical cyclones measuring 16 mm OD at the top and 80 mm from the top of the cyclone body to the bottom of the tapered section. The cyclone inlet and outlet pipes were 2 mm ID and the liquid from the bottom of the cyclone assembly was drained into a 1 liter bottle. The gas displacement tube (10 mm OD) connecting the discharge piston to the cyclone assembly was bent to provide a liquid seal. The atomized vapor from the ketene generator cyclone assembly was then passed to the ketene purification section.

In the ketene cleaning section, the ketene-containing vapor passed through a defogging trap (trap A) maintained at 0ºC to remove a portion of the entrained acetic anhydride/acetic acid. The ketene-containing stream exiting trap A was then directed to a trap (trap B) maintained at -78ºC where the ketene condensed. The outlet line of trap B led to a three-way stopcock (SC1). In one position, SC1 vented excess ketene-containing vapors from trap B to a line connected to the water scrubber via a tee. In another position, SC1 directed the ketene-containing vapors to the ketene inlet line of a ketene trap/evaporator assembly.

The ketene trap/evaporator assembly was a modified two-piece 32 x 200 mm vacuum trap, with the lower part of the trap narrowed to 19 mm OD and extended by a further 100 mm. A 7 mm OD/2 mm ID gas inlet tube extended along the outer body of the ketene trap/evaporator assembly and was connected to the lower part of the extended tube section. The Gas inlet pipe was connected to the nitrogen dosing line L2 which included a stopcock (SC2). The purpose of SC2 was to prevent diffusion of ketene into the line L2 when no nitrogen was flowing through it during the ketene generation process and when the ketene trap/evaporator was being fed. The ketene inlet pipe was the normal 10 mm OD inner pipe found in the standard vacuum trap design. The ketene trap/evaporator drain pipe was the normal 10 mm OD side pipe found in the standard vacuum trap design. The ketene trap/evaporator drain pipe was connected to the nitrogen purge line L3 via a tee. After connecting through a tee to purge the line L3, the ketene trap/evaporator line was connected to a three-way stopcock SC3. Additional connections to SC3 are described later below.

The acetaldehyde purification and feed section contained an acetaldehyde trap/evaporator assembly of identical design to the ketene trap/evaporator assembly, except that it was surrounded by a jacket for liquid coolant, which in turn was enclosed by a vacuum envelope. Line L4 supplied metered nitrogen to the acetaldehyde trap/evaporator gas inlet tube. The acetaldehyde trap/evaporator inlet line allowed the introduction of acetaldehyde distilled from a graduated cylinder. The acetaldehyde trap/evaporator drain line was connected to three-way stopcock SC4. Additional connections to SC4 are described later below.

The reactor-condenser series included a reactor, a reflux condenser, and an acetone/solid carbon dioxide condenser. Details of the reactor design are given in the individual examples. The reactants entered the reactor via line L6. The top of each reactor was plugged into the water-cooled reflux condenser to provide for the return of condensable liquids to the reactor. The top of the reflux condenser was connected to an acetone/solid carbon dioxide condenser. Material condensing in the acetone/solid carbon dioxide condenser was not returned to the reactor but was collected in a trap flask (trap C) attached to the bottom of the condenser. Trap C contained tert-butylhydroquinone (TBHQ), about 40 mg) polymerization inhibitor, and trap C was also kept cooled with solid carbon dioxide. The majority of the vinyl acetate was in trap C along with most of the of unreacted acetaldehyde. Acetic acid evaporating from the reactor was found in trap C. The acetone/solid carbon dioxide cooler drain was connected to the reactor cooler series drain line L7.

The scrubber system included a water scrubber for ketene degradation and an analytical scrubber containing acetic acid for ketene quantification (as acetic anhydride) and for capturing and quantifying vinyl acetate or acetaldehyde not captured by Trap C of the reactor-cooler bank section. The acetic acid (approximately 60 mL) in the analytical scrubber also contained TBHQ (approximately 40 mg) polymerization inhibitor. The analytical scrubber fluid was circulated by a Masterflex® peristaltic pump. A stopcock on the bottom of the analytical scrubber allowed for the removal of the scrubber fluid for analysis and a Claisen adapter on the top of the analytical scrubber allowed for the addition of fresh scrubber fluid. An acetone/solid carbon dioxide cooler was connected to the top of the vertical arm of the Claisen adapter to prevent loss of scrubbing fluids. Access to the scrubbing system was provided by a three-way stopcock SC5. In one position, SC5 sent vapor streams to the water scrubber, and in another position, SC5 sent vapor streams to the analytical scrubber.

The seven main sections were connected by a series of connecting lines, tees and stopcocks. Three-way stopcock SC3 connected the ketene trap/evaporator drain line at the point behind the L3 purge tee to three-way stopcock SC4 via line L5 or to the reactor bypass line. Three-way stopcock SC4 connected line L5 to the acetaldehyde trap/evaporator drain line and to the reactor cooler bank inlet line L6. The bypass line connected to SC3 was connected to the reactor cooler bank drain line L7 via a tee and the other arm of the tee was connected to SC5. SC5 directed the gas flow from the reactor coolant bank vent line L7 or from the bypass line to the water scrubber or to the analysis scrubber as mentioned above. A description of the operation of the device follows.

Acetic anhydride (600 ml/min) and nitrogen (25 standard cubic centimeters per minute, SCCM) were fed via L1 into the ketene generation section for 20 minutes and the product was precipitated in trap B maintained at -78ºC through an acetone/solid carbon dioxide bath. in the ketene purification section. During the ketene generation/condensation process, vent gases leaving trap B were flowed through SC1 to the ketene trap/evaporator assembly and then to the water scrubber via stopcock SC3, bypass line and SC5. Stopcock SC2 was closed during the ketene generation process and the ketene trap/evaporator assembly was maintained at -78°C. During the ketene generation process, the ketene purification process and at all other times when no reactants were flowing to the reactor, a nitrogen purge (25 SCCM) flowed through line L3. The -78°C bath was then removed from trap B and the liquid ketene was then allowed to evaporate and condense in the ketene trap/evaporator assembly. The evaporation process, which lasted approximately 1 hour, yielded approximately 30 mL of pure liquid ketene condensate in the ketene trap/evaporator. The shut-off valve SC1 was then turned over to isolate the ketene trap/evaporator assembly from the ketene generation and purification sections and allowed the vapors leaving the ketene generation and purification sections to flow into the water scrubber.

The acetaldehyde trap/evaporator assembly was loaded by filling reagent grade acetaldehyde into a graduated cylinder and then distilling the acetaldehyde using a warm water bath into the acetaldehyde trap/evaporator assembly maintained at -20ºC. During the acetaldehyde loading procedure, no gas was passed through L4 and the acetaldehyde trap/evaporator drain line was opened to the water scrubber through SC4, SC3 and SC5 with nitrogen (25 SCCM) flowing through L3.

Reactant dosing was started by opening SC2 and feeding nitrogen (88 SCCM) into the ketene trap/vaporizer gas inlet tube through line L2 while maintaining the ketene trap/vaporizer assembly at -78ºC. These conditions provided 0.7 mmol ketene/min. Nitrogen (118 SCCM) was fed into the acetaldehyde trap/vaporizer gas inlet tube through line L4 while maintaining the acetaldehyde trap/vaporizer assembly at -20ºC. These conditions provided 1.0 mmol acetaldehyde/min. Stopcocks SC3 and SC4 were positioned so that all three outlet ports of each stopcock were open and SC5 was positioned to allow venting to the water scrubber. With the stopcocks in this position, the reactants and diluent gases flowed through the bypass line and there was no flow into the reactor because the pressure drop across the reactor was greater than the pressure drop via the bypass line. The purge nitrogen in line L3 was then turned off and access to the bypass line was blocked by turning SC3 to direct the nitrogen/ketene stream to SC4 via line L5 where it mixed with the nitrogen/acetaldehyde stream and continued into the reactor cooler bank via line L6. At this time, stopcock SC5 was positioned to send vapors vented from the reactor cooler bank to the acetic acid analysis scrubber. The reactor heater was also turned on at this time. The reactions were typically run for several hours each day and the time required to heat the reactor (about 20 minutes) was typically short compared to the total time the reactor was heated. The reaction was terminated by opening SC3 to allow all streams to flow into the bypass line, switching SC5 to the water scrubber, stopping nitrogen flow in L2 and L4, closing SC2, reintroducing purge nitrogen into L3, and stopping heat to the reactor. The condensed liquid in trap C of the reactor cooler series was drained, weighed, and analyzed. The acetic acid analysis scrubber was allowed to warm overnight, and the scrubber solution was then drained, weighed, and analyzed. Typically, reactions were run in this manner for several days before the catalyst charge was changed. When the catalyst solution (the reactor residue) was drained from the liquid phase reactor, it was weighed and analyzed. When the catalyst was drained from the reactor containing insoluble solid catalyst, the liquid remaining in the lower part of the reactor (the reactor residue) was weighed and analyzed.

Products obtained from trap C of the reactor-cooler series, acetic acid scrubber solution and reactor residue were analyzed by gas chromatography using Hewlett Packard Model 5890 gas chromatographs using flame ionization detectors. Vinyl acetate, acetaldehyde and acetic acid were analyzed using a 25 m x 0.53 mm FFAP capillary column (1.0 micron film thickness) programmed to 40ºC for 5 minutes, 15ºC/min to 235ºC and hold at 235ºC for 1.67 minutes. Acetic anhydride and ethylidene diacetate were analyzed using a 30 m x 0.53 mm DB-5 capillary column (1.5 micron film thickness) programmed to 40ºC for 8 minutes, 7ºC/min to 200ºC with a hold time of 0 minutes at 200ºC. Mixtures were prepared for gas chromatography analysis by adding 5 ml of a tetrahydrofuran solution containing 2% Decane from the internal standard was added to an accurately weighed 1 g sample of the reaction product.

In the examples below, the percent yield of vinyl acetate is defined as 100 times the moles of vinyl acetate produced divided by the moles of ketene fed.

example 1

This example illustrates the process of the invention conducted with the reactor zone containing acetic anhydride solvent and p-toluenesulfonic acid catalyst at 150°C and a pressure of 1 bar (100 kPa) absolute. The example also illustrates conditions which provide high yields of vinyl acetate without the production of acetic acid after a steady state is reached (Examples 1.2 to 1.4). The reactor was a 36 mm OD x 175 mm long tube sealed at one end. The reactor was mounted vertically with the open end upward. The reactor was charged with acetic anhydride (65.8 g, 0.645 mol), p-toluenesulfonic acid monohydrate (5.8 g, 0.0306 mol) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reaction tube was placed in a temperature controlled electric heater. A three-way Claisen adapter was connected to the open end of the reactor. A 10 mm OD gas supply tube with a lower end extended to 1 mm ID was inserted through the straight vertical arm of the Claisen adapter and extended to within about 1 mm of the reactor base. A threaded transition pipe piece at the upper end of the vertical arm of the Claisen adapter held the gas supply tube in place and provided a seal. The reactor inlet line L6 was connected to the gas supply tube. The curved side arm of the Claisen adapter was connected to the base of the reflux condenser portion of the condenser bank assembly. To start the reaction of Example No. 1.1, the stopcock SC3 was positioned to block access to the bypass line, thereby passing ketene (0.7 mmol/min), acetaldehyde (1.0 mmol/min) and nitrogen (206 SCCM) to the reactor through the gas feed tube, with the gas leaving the reactor contacting the water cooled reflux condenser, the solid carbon dioxide/acetone cooled condenser and the trap flask containing TBHQ (40 mg, 0.24 mmol), and SC5 was positioned so that the product leaving the solid carbon dioxide/acetone condenser was passed to the acetic acid (the also containing about 40 mg of TBHQ) scrubber and the heater temperature was set at 150°C. The 150°C reactor temperature was reached after 20 minutes. The reaction was allowed to continue in this manner for a total of 314 minutes. To terminate the reaction, SC3 was positioned to bypass the reactant gases through the reactor bypass line, heating of the reactor was stopped, and SC5 was positioned to direct all gases to the water scrubber. The gas supply was stopped and the reactor system was given access to nitrogen (25 SCCM) flowing through line L3 during the cool down period and was idle at all other times. The liquid contained in trap C was weighed and analyzed by gas chromatography. The acetic acid scrubber was allowed to warm to room temperature and drain overnight. The product recovered from the acetic acid scrubber was weighed and analyzed by gas chromatography. The reactor residue solution was retained in the reactor. Examples Nos. 1.2, 1.3 and 1.4 were carried out by conducting the reaction in the manner of Example 1.1 for periods of 301, 300 and 300 minutes respectively using the same reactor residue solution. Following are the yields of vinyl acetate isolated from the cooler series trap C and the acetic acid scrubber solutions and the wt% of acetic acid contained in trap C (trap C wt% HOAc).

The reactor residue solution remaining from Example 1.4 was weighed and analyzed by gas chromatography. The recovered residue solution (86.73 g) contained acetaldehyde (0.31 wt%), vinyl acetate (1.7 wt%), acetic acid (1.14 wt%), acetic anhydride (57.32 wt%) and EDA (28.99 wt%).

Example 2

This example illustrates the effect of operating the process of Example 1 at a lower temperature. The example also illustrates conditions required for high yields of vinyl acetate without the generation of acetic acid after a steady state has been reached (Examples 2.2 to 2.5). The same reactor system as used in Example 1 was used in Example 2. The reaction tube was charged with acetic anhydride (65.8 g, 0.645 mol), p-toluenesulfonic acid monohydrate (5.8 g, 0.0306 mol) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor as in Example 1. The reactor was operated in the same manner as in Example 1 except that the temperature was set at 140°C. Examples Nos. 2.1, 2.2, 2.3, 2.4 and 2.5 were each run for 300 minutes using the same residue solution. The yields of vinyl acetate and acetic acid scrubbing solutions isolated from Trap C and the wt% acetic acid contained in Trap C (Trap C wt% HOAc) are given below.

The reactor residue solution obtained in Example 2.5 was analyzed by gas chromatography. The recovered residue solution (88.18 g) contained acetaldehyde (0.41 wt%), vinyl acetate (1.59 wt%), acetic acid (1.03 wt%), acetic anhydride (47.65 wt%), and EDA (39.42 wt%).

Example 3

This example illustrates the effect of operating the process of Example 1 at a higher temperature. The example also illustrates conditions that provide high yields of vinyl acetate without the production of acetic acid after steady state is achieved (Examples 3.2 through 3.5). The same reactor system as used in Example 1 was used in Example 3. The reaction tube was charged with acetic anhydride (65.8 g, 0.645 mole), p-toluenesulfonic acid monohydrate (5.8 g, 0.0306 mole) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor as in Example 1. The reactor was operated in the same manner as in Example 1 except that the temperature was set at 160°C. Examples Nos. 3.1, 3.2, 3.3, 3.4 and 3.5 were each heated for 300 minutes under using the same residual solution. The yields of vinyl acetate and acetic acid scrubbing solutions isolated from Trap C and the wt% acetic acid contained in Trap C (Trap C wt% HOAc) are given below.

The reactor residue solution obtained in Example 3.5 was analyzed by gas chromatography. The recovered residue solution (86.31 g) contained acetaldehyde (0.33 wt%), vinyl acetate (1.41 wt%), acetic acid (1.23 wt%), acetic anhydride (56.12 wt%), and EDA (28.82 wt%).

Example 4

This example illustrates the process of the invention accomplished with the reactor zone containing N-methyl-2-pyrrolidinone solvent and p-toluenesulfonic acid catalyst at 150°C and a pressure of 1 bar (100 kPa) absolute. The example also illustrates the conditions that provide high yields of vinyl acetate without the production of acetic acid after steady state is reached (Examples 4.2 to 4.5). The same reactor system as used in Example 1 was used in Example 4. The reaction tube was charged with N-methyl-2-pyrrolidinone (63.2 g, 0.638 mol), p-toluenesulfonic acid monohydrate (5.8 g, 0.0306 mol) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reactor was operated in the same manner as in Example 1 at 150°C. Examples Nos. 4.1, 4.2, 4.3, 4.4 and 4.5 were run for 300, 330, 300, 305 and 300 minutes respectively using the same residue solution. The yields of vinyl acetate isolated from Trap C and the acetic acid scrubbing solutions and the wt% acetic acid contained in Trap C (Trap C wt% HOAc) follow below.

The reactor residue solution obtained in Example 4.5 was analyzed by gas chromatography. The recovered residue solution (80.81 g) contained acetaldehyde (1.04 wt%), acetic acid (1.01 wt%), acetic anhydride (4.73 wt%) and a small amount of EDA (1.13 wt%).

Example 5

This example illustrates the process of the invention which was accomplished with the reactor zone containing mixed acetic anhydride-acetic acid solvent and p-toluenesulfonic acid catalyst at 150°C and at a pressure of 1 bar (100 kPa) absolute. The example also illustrates the conditions which provide high yields of vinyl acetate without the production of acetic acid after steady state is reached (Examples 5.2 to 5.4). The same reactor system as used in Example 1 was used in Example 5. The reaction tube was charged with acetic anhydride (59.5 g, 0.583 mole), acetic acid (3.7 g, 0.0616 mole), p-toluenesulfonic acid monohydrate (5.8 g, 0.0306 mole) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reactor was operated in the same manner as in Example 1 at 150°C. Examples Nos. 5.1, 5.2, 5.3 and 5.4 were run for 340, 300, 300 and 300 minutes respectively using the same residue solution. The yields of vinyl acetate and acetic acid scrubbing solutions isolated from Trap C and the wt% acetic acid contained in Trap C (Trap C wt% HOAc) follow below.

The reactor residue solution obtained in Example 5.4 was analyzed by gas chromatography. The recovered residue solution (80.68 g) contained acetaldehyde (0.36 wt%), vinyl acetate (0.71 wt%), acetic anhydride (62.53 wt%) and a small amount of EDA (35.7 wt%).

Example 6

This example illustrates the process of the invention accomplished with the reactor zone containing mixed acetic anhydride-acetic acid solvent and at 150°C as in Example 5, but with benzenesulfonic acid catalyst instead. The example also shows that the stronger acid catalyst continues to produce acetic acid after the vinyl acetate yield has been maximized (Examples 6.2 to 6.4). The same reactor system as used in Example 1 was used in Example 6. The reaction tube was charged with acetic anhydride (59.5 g, 0.583 mol), acetic acid (3.7 g, 0.0616 mol), benzenesulfonic acid monohydrate (5.4 g, 0.0306 mol), and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reactor was operated in the same manner as in Example 1 at 150°C. Examples Nos. 6.1, 6.2, 6.3 and 6.4 were run for 305, 300, 300 and 300 minutes respectively using the same residue solution. The yields of vinyl acetate and acetic acid scrubbing solutions isolated from Trap C and the wt% acetic acid contained in Trap C (Trap C wt% HOAc) follow below.

The reactor residue solution obtained in Example 6.4 was analyzed by gas chromatography. The recovered residue solution (83.24 g) contained vinyl acetate (0.55 wt%), acetic acid (3.18 wt%), acetic anhydride (58.7 wt%) and EDA (33.07 wt%).

Example 7

This example illustrates the process of the invention accomplished with the reactor zone containing mixed acetic anhydride-acetic acid solvent and at 150°C as in Example 6, but with methanesulfonic acid catalyst instead. The example also shows that no acetic acid is formed when a weaker acid is used, but that the vinyl acetate yield is reduced. The same reactor system as used in Example 1 was used in Example 7. The reaction tube was charged with acetic anhydride (59.5 g, 0.583 mole), acetic acid (3.7 g, 0.0616 mole), methanesulfonic acid (2.9 g, 0.0306 mole), and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reactor was operated in the same manner as in Example 1 at 150°C. Examples Nos. 7.1, 7.2 and 7.3 were run for 300, 300 and 305 minutes respectively using the same residue solution. The yields of vinyl acetate and acetic acid scrubbing solutions isolated from Trap C and the wt% acetic acid contained in Trap C (Trap C wt% HOAc) follow below.

The reactor residue solution obtained in Example 7.3 was analyzed by gas chromatography. The recovered residue solution (75.31 g) contained vinyl acetate (0.91 wt%), acetic acid (1.22 wt%), acetic anhydride (79.9 wt%) and EDA (9.69 wt%).

Example 8

This example illustrates the process of the invention which was accomplished with the reactor zone containing mixed acetic anhydride-acetic acid solvent and at 150°C as in Example 7, but with sulfuric acid catalyst instead. The example also shows that strong acids reduce the vinyl acetate yield and form acetic acid. The same reactor system as used in Example 1 was used in Example 8. The reaction tube was charged with acetic anhydride (59.5 g, 0.583 mole), acetic acid (3.7 g, 0.0616 mole), 96.8% sulfuric acid (3.1 g, 0.0306 mole) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reactor was operated in the same manner as in Example 1 at 150°C for 300 minutes. The yield of vinyl acetate and acetic acid scrubber isolated from trap C was 12%, and trap C contained 19.28 wt.% acetic acid. Clogging of the gas feed pipe due to excessive char generation prevented further operation of the reactor containing this residue solution. The liquid portion (58.28 g) of the residue mixture was recovered and contained acetic acid (21.14 wt.%), acetic anhydride (55.83 wt.%) and EDA (24.95 wt.%).

Example 9

This example illustrates the process of the invention accomplished with the reactor zone containing mixed acetic anhydride-acetic acid solvent and at 150°C as in Example 8, but with phosphoric acid catalyst instead. The example also demonstrates that strong acids reduce vinyl acetate yield and form acetic acid. The same reactor system as used in Example 1 was used in Example 9. The reaction tube was charged with acetic anhydride (59.5 g, 0.583 mole), acetic acid (3.7 g, 0.0616 mole), 85% phosphoric acid (3.53 g, 0.0306 mole) and TBHQ (140 mg, 0.84 mmol) polymerization inhibitor. The reactor was operated in the same manner as in Example 1 at 150°C for 300 minutes. The yield of vinyl acetate and acetic acid scrubber isolated from Trap C was 19%, and Trap C contained 13.84 wt% acetic acid. Plugging of the gas feed tube due to excessive char generation allowed the reactor containing this residue solution to continue operating for only 65 minutes. The liquid portion (72.25 g) of the residue mixture was recovered and contained vinyl acetate (0.84 wt%), acetic acid (1.98 wt%), acetic anhydride (77.55 wt%), and EDA (32.78 wt%).

Example 10

This example illustrates the process of the invention wherein the reactor zone contains the insoluble solid acid catalyst Amberlyst® 15 and no solvent is added. The glass reactor used in this example consisted of a 74 cm x 25 mm OD tube equipped with a permanent temperature probe extending from the base of the reactor. The middle section of the reaction tube was constructed with a condenser jacket which in turn was surrounded by a vacuum envelope to to prevent heat loss. The length of the jacketed portion was 61 cm. The 25 mm OD tube had notches 5 cm above the base of the jacket to support the catalyst bed. A physical mixture of Amberlyst® 15 resin (17 mL, 10.26 g, ca. 31 meq -SO₃H) and 16 x 24 mesh quartz chips (50 mL) was prepared. The lower portion of the reactor was filled with 6 x 6 mm Raschig rings to a height of 6 cm above the notches and a 1 cm layer of 4 x 8 mesh quartz chips was added on top of the Raschig rings. The entire mixture of Amberlyst® 15 and 16 x 24 mesh quartz chips was then added to the reactor. The length of the catalyst + quartz mixture charge was 20 cm. An additional 5 cm layer of 4 x 8 mesh quartz chips was placed on top of the catalyst bed. A 100 mL double-necked flask was attached to the bottom of the reactor. The reactor inlet line L6 was connected to the other neck of the 100 mL double-necked flask. The bottom of the reflux condenser section condenser bank was connected to the top of the reactor to allow the return of condensable liquids to the reactor. The 100 mL double-necked flask served as a residue reservoir for liquid effluent from the reactor. Under the conditions of the example, no liquid flowed into the residue reservoir until the reaction feed streams to the reactor were completed. Ketene, acetaldehyde and nitrogen were fed to the reactor at the start of Example 10.1 at a pressure of about 1 bar (100 kPa) absolute at the same rate as for Example 1. The catalyst bed temperature rose slowly from 19.6°C to 83.3°C, whereupon steam was released into the reactor cooler jacket. The catalyst bed temperature continued to rise slowly to 110.6°C and then slowly decreased to level off at 98.7°C. During the process of Example 10.1, the catalyst bed slowly became wet. The reaction was terminated after 360 minutes for Example 1, whereupon a portion of the catalyst bed wetting liquid was drained to the reservoir. In Examples 10.2 to 10.5, the reactor was operated in the same manner as in Example 10.1, except that steam heating of the reactor was initiated when the charges were initially introduced into the reactor. Examples 10.2, 10.3, 10.4 and 10.5 were run for periods of 300, 330, 300 and 352 minutes, respectively, using the same catalyst bed. The yields of vinyl acetate isolated from Trap C and the acetic acid scrubbing solutions and the wt. % of acetic acid contained in Trap C (Trap C wt. HOAc) are reported below.

The reactor residue solution recovered from the reservoir after completion of the reaction of Example 10.5 was analyzed by gas chromatography. The recovered residue solution (8.1 g) contained acetaldehyde (3.46 wt%), acetic acid (1.17 wt%), acetic anhydride (2.01 wt%) and EDA (39.9 wt%).

Example 11

This example illustrates the process of the invention using a single contact zone for the production of vinyl acetate from ketene and hydrogen. The same reactor system as used in Example 1 was used in Example 11, but the feed system was modified by shutting off the acetaldehyde feed system and replacing the nitrogen feed to the ketene trap/evaporator section with a hydrogen stream. The analytical scrubber was charged with methanol instead of acetic acid. The reactor was charged with 5% Pd on carbon powder (2.0151 g), acetic anhydride (65.8 g, 0.645 mol), p-toluenesulfonic acid monohydrate (5.8 g, 0.0306 mol), and TBHQ (160 mg, 0.963 mmol). The reactor was operated as in Example 1 at 140°C for 317 minutes, but the feed consisted only of ketene (0.7 mmol/min) and hydrogen (88 SCCM). The material isolated from trap C (4.32 g) contained acetaldehyde (11.2 wt%), vinyl acetate (26.7 wt%), acetic acid (37.7 wt%), and EDA (2.1 wt%). The material isolated from the methanol scrubber (76.6 g) contained methyl acetate (0.95 wt%).

Claims (14)

1. A process for producing vinyl acetate, comprising the steps 1) contacting, in a first contact zone, hydrogen and ketene gases with a catalyst comprising a metal selected from the elements of groups 9 and 10 of the Periodic Table; 2) recovering a product comprising acetaldehyde from the first contact zone; 3) contacting, in a second contact zone, a mixture of ketene and the acetaldehyde obtained from the first contact zone with an acid catalyst; and 4) recovering a product comprising vinyl acetate from the second contact zone. 2. A process according to claim 1, wherein the contacting in the first contact zone is carried out at 590 to 200°C and at a pressure of 0.1 to 20 bar (10 to 2000 kPa) absolute, and the contacting in the second contact zone is carried out at 85 to 200°C and at 0.05 to 20 bar (5 to 2000 kPa) absolute. 3 A method according to claim 2, wherein a non-reactive diluent gas is also fed into at least one of the contact zones. 4. A process according to claim 2, wherein the contacting in the first contact zone is carried out at 70° to 150°C and at 0.25 to 10 bar (25 to 1000 kPa) absolute and the contacting in the second contact zone is carried out at 100 to 180°C and at 0.1 to 5 bar (10 to 500 kPa) absolute. 5. The process of claim 1, wherein the acid catalyst in the second contact zone comprises a Bronsted acid. 6. The method of claim 5, wherein the first contact zone further contains a non-reactive liquid. 7. The method of claim 5, wherein the second contact zone further contains a solvent. 8. The process of claim 1, wherein the catalyst is a supported or unsupported palladium catalyst. 9. The process of claim 5, wherein the Bronsted acid is selected from the group consisting of phosphoric acid, sulfuric acid, methanesulfonic acid, benzenesulfonic acid, p-toluenesulfonic acid, naphthalenesulfonic acid, naphthalenedisulfonic acid and an acidic ion exchange resin. 10. The process of claim 9, wherein the Bronsted acid is p-toluenesulfonic acid. 11. A process for producing vinyl acetate comprising the steps of 1) contacting ketene and hydrogen gases with a hydrogenation catalyst comprising a metal selected from the elements of Groups 9 and 10 of the Periodic Table and an acid catalyst in a contact zone; and 2) withdrawing a product comprising vinyl acetate from the contact zone. 12. The process of claim 11, wherein the contact zone has a temperature of 100 to 180°C and a pressure of 0.1 to 10 bar (10 to 1000 kPa) absolute. 13. The process of claim 12, wherein the hydrogenation catalyst is supported or unsupported palladium catalyst and the acid catalyst comprises a Bronsted acid. 14. The process of claim 13, wherein the Bronsted acid is p-toluenesulfonic acid.